Viewing Issue Advanced Details
ID Category [?] Severity [?] Reproducibility Date Submitted Last Update
07358 Media Support Critical (emulation) Always Jun 13, 2019, 12:22 Jun 14, 2019, 00:12
Tester MrGW View Status Public Platform MAME (Self-compiled)
Assigned To Resolution Invalid report OS Linux (64-bit)
Status [?] Closed Driver coco3.cpp
Version 0.210 Fixed in Version Build 64-bit
Summary MESS-specific 07358: coco3, coco12: MAME crashes with segfault when accessing floppy drive image
Description MAME will crash (segfault) when accessing floppy drive image. It can be using any disk commands (i.e. DIR, DSKINI, etc.).

The same issue happens when using the coco3 and coco12 drivers.
Steps To Reproduce Use the following command line:

mame coco3 -flop1 TEST.DSK

Once MAME is loaded, type DIR
Additional Information I'm using Linux Mint 18.3

gcc (Ubuntu 8.1.0-5ubuntu1~16.04) 8.1.0
Copyright (C) 2018 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

g++ (Ubuntu 8.1.0-5ubuntu1~16.04) 8.1.0
Copyright (C) 2018 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

ldd (Ubuntu GLIBC 2.23-0ubuntu11) 2.23
Copyright (C) 2016 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
Written by Roland McGrath and Ulrich Drepper.

I've used the same process for compiling and have not had any issues for quite a few releases. Recompiling MAME 0.209 eliminates the issue.

I do not see the issue when using the official Windows binary image.
Flags
Regression Version
Affected Sets / Systems coco3, coco12
Attached Files
zip file icon mame-coco3.zip (1,693 bytes) Jun 13, 2019, 12:22 Uploaded by MrGW
gdb log and TEST.DSK floppy image
Relationships
There are no relationship linked to this issue.
Notes
3
User avatar
No.16536
Tafoid
Administrator
Jun 13, 2019, 16:28
edited on: Jun 13, 2019, 16:29
We had this issue/regression at the very beginning of the 0.211 development cycle we are in now, so you've likely compiled code in MAME that is after the initial source cutoff date and before the fix was put in days later.

Compare your code against:
https://github.com/mamedev/mame/releases/tag/mame0210 (includes initial code up to release0.210)

Initial Source Freeze happened after a source clean:
May 26st - Branch 'release0.210" based on source up to 626b566feeea0863644a45232b92568417542af9

Bug introduced (after initial source freeze):
May 28th - https://github.com/mamedev/mame/commit/9863c37e286cef5924cf22980276f64aa42bed2b

Fixed after the mame0.210 tag was produced, but even the breakage was not in 0.210 sources:
May 31st - https://github.com/mamedev/mame/commit/588a5c397f3d890bea9e679ab816e4b10855b2a7

Apologies if this is confusing.

Typically the Saturday/Sunday before the expected end of month Wednesday release - the source is frozen for the upcoming release, source is made into a "release0xxx" branch. This means that any work afterwards will be applied to the next months' release... save for a few 'cherry picked' bugfixes or important and non-threatening code additions. This isn't a bug in a released version so it isn't a valid bug report.

Since it has been fixed, a current pull of GIT will correct it, or a proper pull of the mame0210 tag at the link above will give you the proper sources needed to mirror what is released on mainline Windows. Please be sure you are using the proper sources before reporting regression. Technically, this regression didn't happen until 0.211 WIP code, so there are no integer releases with this issue.

I'll leave until I can get confirmation on your end that you understand why this regression only happens in your compile since your source code has the added (0.211 WIP) changes.
User avatar
No.16537
MrGW
Tester
Jun 13, 2019, 18:32
Hi Tafoid,

This makes sense and I'll pull a current copy from git, compile and re-test.

Thanks for the thorough explanation!
User avatar
No.16539
MrGW
Tester
Jun 13, 2019, 22:29
Compiled from current source. Issue is resolved.

Thank you!